Male youth in urban environments bear a disproportionate burden of violence perpetration and victimization, which can negatively impact physical and mental health. Strengthening adolescent-adult connections is a promising evidence-based strategy to increase social support and prevent youth violence. However, studies suggest that the simple presence of a supportive adult may not universally confer protection in low-resource urban neighborhoods, and that concurrent social influences may dictate risk. The overall goal of this proposal is to apply social network analysis techniques to elucidate the structure and quality of adolescents' adult support networks across family, school, and community contexts that confer protection from violence in low- resource urban neighborhoods. Results will inform the design and pilot testing of a social network-based adaptation of the `building intergenerational partnerships' component of Youth Empowerment Solutions (YES), an evidence-based violence prevention program that fosters community engagement and reduces aggressive behavior. We will leverage data, research infrastructure, and community partnerships from an ongoing study engaged with male adolescents in twenty disadvantaged neighborhoods in Pittsburgh, PA. In Aim 1, we will identify adolescent-adult support network structures and relationship qualities across family, school, and community contexts inversely related to violence involvement and explore mechanisms through which these networks confer protection. In Aim 2, we will design and iteratively test a stakeholder-engaged community- based social network adaptation of the YES intervention designed to strengthen and expand adolescent-adult support networks. In Aim 3, we will conduct a pilot cluster-randomized feasibility trial of the adapted YES intervention among adolescent males and key adult supports. Completion of these aims will provide critical insights into how adolescent-adult support networks can be leveraged to reduce youth violence in urban environments. The proposal is supported by an interdisciplinary mentorship team with nationally-recognized expertise in youth violence, social network analysis, intervention design, and clinical trial implementation. With guidance from committed mentors and strong institutional support, the PI will receive the necessary intensive mentorship, didactic education, and research experience to become an independent investigator, supporting career development objectives in social network analysis, intervention design, and implementation science. This work will provide essential preliminary data to support future grant applications to: 1) use social network analysis methods to investigate how multimodal interactions across social and environmental contexts affect violence risk, and 2) to test in a cluster randomized controlled trial the effectiveness of a social network-based violence prevention program. The PI will be poised to become an independent physician-scientist leading studies aligned with NICHD's Pediatric Trauma and Critical Illness Branch priorities to support collaborative multidisciplinary youth violence prevention and intervention research.